文献速递肺结节与肺癌

自然语言处理识别肺结节并提取结节特征

Natural Language Processing to Identify Pulmonary Nodules and Extract Nodule Characteristics From Radiology Reports.
PubMed 文献 · Chest · 2021-06-04

该研究修订并扩展了一个自然语言处理(NLP)算法,用于从放射学报告中识别肺结节并提取最大结节的特征(大小、肺叶、侧别、衰减、钙化、边缘)。在200份随机测试样本中,NLP识别肺结节的灵敏度为98.6%(95%CI 95.0%-99.8%),特异度为100%(95%CI 93.9%-100%),正确识别最大结节直径的比例为95.7%(95%CI 91.0%-98.4%)。在717,304份胸部CT报告中,NLP发现217,771份(30.4%)有结节,2006至2016年结节报告数增加150%,最大结节平均大小从11 mm降至8.9 mm。

要点

  • NLP算法识别肺结节的灵敏度98.6%、特异度100%,最大结节直径识别准确率95.7%。
  • 在717,304份胸部CT报告中,NLP检出30.4%有结节,2006-2016年结节报告数增加150%。
  • 放射科医生对侧别和肺叶的记录率(90%-95%)高于衰减、钙化和边缘特征(11%-14%)。

核心结论

  • NLP算法能以高准确率识别肺结节并提取相关特征。
  • 社区实践中结节特征记录不完整,需改进记录。
  • NLP算法可用于基于人群的研究,避免耗力的病历审查。

研究设计与样本

  • 研究修订并扩展了现有NLP算法,用于识别放射学报告中的肺结节及最大结节特征。
  • 在200份随机放射学报告测试样本中,将NLP结果与人工审查参考标准比较。
  • 将最终自动化方法应用于2006至2016年综合医疗系统中接受胸部CT扫描的较大患者队列,描述其人口学和临床特征。

主要发现

  • 测试样本中,NLP识别肺结节的灵敏度为98.6%(95%CI 95.0%-99.8%),特异度为100%(95%CI 93.9%-100%)。
  • 对于衰减、边缘和钙化,NLP达到相似准确度;正确识别最大结节直径141例中135例(95.7%;95%CI 91.0%-98.4%)。
  • 较大队列中,717,304份胸部CT报告里NLP发现217,771份有结节(30.4%)。
  • 2006至2016年,有结节的报告数增加150%,最大结节平均大小从11 mm逐渐降至8.9 mm。
  • 放射科医生记录侧别和肺叶(90%-95%)的频率高于衰减、钙化和边缘特征(11%-14%)。

关键数值与推荐

  • NLP识别肺结节灵敏度98.6%(95%CI 95.0%-99.8%),特异度100%(95%CI 93.9%-100%)。
  • 最大结节直径识别准确率95.7%(95%CI 91.0%-98.4%)。
  • 结节报告占比30.4%(217,771/717,304)。
  • 2006至2016年结节报告数增加150%,最大结节平均大小从11 mm降至8.9 mm。

临床怎么用

  • NLP算法可用于人群研究,自动识别肺结节,替代人工病历审查。
  • 临床中应改进结节特征(衰减、钙化、边缘)的记录完整性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: There is an urgent need for population-based studies on managing patients with pulmonary nodules. RESEARCH QUESTION: Is it possible to identify pulmonary nodules and associated characteristics using an automated method? STUDY DESIGN AND METHODS: We revised and refined an existing natural language processing (NLP) algorithm to identify radiology transcripts with pulmonary nodules and greatly expanded its functionality to identify the characteristics of the largest nodule, when present, including size, lobe, laterality, attenuation, calcification, and edge. We compared NLP results with a reference standard of manual transcript review in a random test sample of 200 radiology transcripts. We applied the final automated method to a larger cohort of patients who underwent chest CT scan in an integrated health care system from 2006 to 2016, and described their demographic and clinical characteristics. RESULTS: In the test sample, the NLP algorithm had very high sensitivity (98.6%; 95% CI, 95.0%-99.8%) and specificity (100%; 95% CI, 93.9%-100%) for identifying pulmonary nodules. For attenuation, edge, and calcification, the NLP algorithm achieved similar accuracies, and it correctly identified the diameter of the largest nodule in 135 of 141 cases (95.7%; 95% CI, 91.0%-98.4%). In the larger cohort, the NLP found 217,771 reports with nodules among 717,304 chest CT reports (30.4%). From 2006 to 2016, the number of reports with nodules increased by 150%, and the mean size of the largest nodule gradually decreased from 11 to 8.9 mm. Radiologists documented the laterality and lobe (90%-95%) more often than the attenuation, calcification, and edge characteristics (11%-14%). INTERPRETATION: The NLP algorithm identified pulmonary nodules and associated characteristics with high accuracy. In our community practice settings, the documentation of nodule characteristics is incomplete. Our results call for better documentation of nodule findings. The NLP algorithm can be used in population-based studies to identify pulmonary nodules, avoiding labor-intensive chart review.

来源:Chest,查看原文。版权归原作者所有。

引用

  1. Zheng C, Huang BZ, Agazaryan AA, et al. Natural Language Processing to Identify Pulmonary Nodules and Extract Nodule Characteristics From Radiology Reports. Chest. 2021. PMID: 34089738. doi:10.1016/j.chest.2021.05.048

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。